/
predictor.html
731 lines (666 loc) · 37.2 KB
/
predictor.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1" />
<meta name="generator" content="pdoc 0.9.2" />
<title>ktrain.text.predictor API documentation</title>
<meta name="description" content="" />
<link rel="preload stylesheet" as="style" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/11.0.1/sanitize.min.css" integrity="sha256-PK9q560IAAa6WVRRh76LtCaI8pjTJ2z11v0miyNNjrs=" crossorigin>
<link rel="preload stylesheet" as="style" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/11.0.1/typography.min.css" integrity="sha256-7l/o7C8jubJiy74VsKTidCy1yBkRtiUGbVkYBylBqUg=" crossorigin>
<link rel="stylesheet preload" as="style" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/10.1.1/styles/github.min.css" crossorigin>
<style>:root{--highlight-color:#fe9}.flex{display:flex !important}body{line-height:1.5em}#content{padding:20px}#sidebar{padding:30px;overflow:hidden}#sidebar > *:last-child{margin-bottom:2cm}.http-server-breadcrumbs{font-size:130%;margin:0 0 15px 0}#footer{font-size:.75em;padding:5px 30px;border-top:1px solid #ddd;text-align:right}#footer p{margin:0 0 0 1em;display:inline-block}#footer p:last-child{margin-right:30px}h1,h2,h3,h4,h5{font-weight:300}h1{font-size:2.5em;line-height:1.1em}h2{font-size:1.75em;margin:1em 0 .50em 0}h3{font-size:1.4em;margin:25px 0 10px 0}h4{margin:0;font-size:105%}h1:target,h2:target,h3:target,h4:target,h5:target,h6:target{background:var(--highlight-color);padding:.2em 0}a{color:#058;text-decoration:none;transition:color .3s ease-in-out}a:hover{color:#e82}.title code{font-weight:bold}h2[id^="header-"]{margin-top:2em}.ident{color:#900}pre code{background:#f8f8f8;font-size:.8em;line-height:1.4em}code{background:#f2f2f1;padding:1px 4px;overflow-wrap:break-word}h1 code{background:transparent}pre{background:#f8f8f8;border:0;border-top:1px solid #ccc;border-bottom:1px solid #ccc;margin:1em 0;padding:1ex}#http-server-module-list{display:flex;flex-flow:column}#http-server-module-list div{display:flex}#http-server-module-list dt{min-width:10%}#http-server-module-list p{margin-top:0}.toc ul,#index{list-style-type:none;margin:0;padding:0}#index code{background:transparent}#index h3{border-bottom:1px solid #ddd}#index ul{padding:0}#index h4{margin-top:.6em;font-weight:bold}@media (min-width:200ex){#index .two-column{column-count:2}}@media (min-width:300ex){#index .two-column{column-count:3}}dl{margin-bottom:2em}dl dl:last-child{margin-bottom:4em}dd{margin:0 0 1em 3em}#header-classes + dl > dd{margin-bottom:3em}dd dd{margin-left:2em}dd p{margin:10px 0}.name{background:#eee;font-weight:bold;font-size:.85em;padding:5px 10px;display:inline-block;min-width:40%}.name:hover{background:#e0e0e0}dt:target .name{background:var(--highlight-color)}.name > span:first-child{white-space:nowrap}.name.class > span:nth-child(2){margin-left:.4em}.inherited{color:#999;border-left:5px solid #eee;padding-left:1em}.inheritance em{font-style:normal;font-weight:bold}.desc h2{font-weight:400;font-size:1.25em}.desc h3{font-size:1em}.desc dt code{background:inherit}.source summary,.git-link-div{color:#666;text-align:right;font-weight:400;font-size:.8em;text-transform:uppercase}.source summary > *{white-space:nowrap;cursor:pointer}.git-link{color:inherit;margin-left:1em}.source pre{max-height:500px;overflow:auto;margin:0}.source pre code{font-size:12px;overflow:visible}.hlist{list-style:none}.hlist li{display:inline}.hlist li:after{content:',\2002'}.hlist li:last-child:after{content:none}.hlist .hlist{display:inline;padding-left:1em}img{max-width:100%}td{padding:0 .5em}.admonition{padding:.1em .5em;margin-bottom:1em}.admonition-title{font-weight:bold}.admonition.note,.admonition.info,.admonition.important{background:#aef}.admonition.todo,.admonition.versionadded,.admonition.tip,.admonition.hint{background:#dfd}.admonition.warning,.admonition.versionchanged,.admonition.deprecated{background:#fd4}.admonition.error,.admonition.danger,.admonition.caution{background:lightpink}</style>
<style media="screen and (min-width: 700px)">@media screen and (min-width:700px){#sidebar{width:30%;height:100vh;overflow:auto;position:sticky;top:0}#content{width:70%;max-width:100ch;padding:3em 4em;border-left:1px solid #ddd}pre code{font-size:1em}.item .name{font-size:1em}main{display:flex;flex-direction:row-reverse;justify-content:flex-end}.toc ul ul,#index ul{padding-left:1.5em}.toc > ul > li{margin-top:.5em}}</style>
<style media="print">@media print{#sidebar h1{page-break-before:always}.source{display:none}}@media print{*{background:transparent !important;color:#000 !important;box-shadow:none !important;text-shadow:none !important}a[href]:after{content:" (" attr(href) ")";font-size:90%}a[href][title]:after{content:none}abbr[title]:after{content:" (" attr(title) ")"}.ir a:after,a[href^="javascript:"]:after,a[href^="#"]:after{content:""}pre,blockquote{border:1px solid #999;page-break-inside:avoid}thead{display:table-header-group}tr,img{page-break-inside:avoid}img{max-width:100% !important}@page{margin:0.5cm}p,h2,h3{orphans:3;widows:3}h1,h2,h3,h4,h5,h6{page-break-after:avoid}}</style>
<script defer src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/10.1.1/highlight.min.js" integrity="sha256-Uv3H6lx7dJmRfRvH8TH6kJD1TSK1aFcwgx+mdg3epi8=" crossorigin></script>
<script>window.addEventListener('DOMContentLoaded', () => hljs.initHighlighting())</script>
</head>
<body>
<main>
<article id="content">
<header>
<h1 class="title">Module <code>ktrain.text.predictor</code></h1>
</header>
<section id="section-intro">
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">from ..imports import *
from ..predictor import Predictor
from .preprocessor import TextPreprocessor, TransformersPreprocessor, detect_text_format
from .. import utils as U
class TextPredictor(Predictor):
"""
```
predicts text classes
```
"""
def __init__(self, model, preproc, batch_size=U.DEFAULT_BS):
if not isinstance(model, Model):
raise ValueError('model must be of instance Model')
if not isinstance(preproc, TextPreprocessor):
#if type(preproc).__name__ != 'TextPreprocessor':
raise ValueError('preproc must be a TextPreprocessor object')
self.model = model
self.preproc = preproc
self.c = self.preproc.get_classes()
self.batch_size = batch_size
def get_classes(self):
return self.c
def predict(self, texts, return_proba=False):
"""
```
Makes predictions for a list of strings where each string is a document
or text snippet.
If return_proba is True, returns probabilities of each class.
Args:
texts(str|list): For text classification, texts should be either a str or
a list of str.
For sentence pair classification, texts should be either
a tuple of form (str, str) or list of tuples.
A single tuple of the form (str, str) is automatically treated as sentence pair classification, so
please refrain from using tuples for text classification tasks.
return_proba(bool): If True, return probabilities instead of predicted class labels
```
"""
is_array, is_pair = detect_text_format(texts)
if not is_array: texts = [texts]
classification, multilabel = U.is_classifier(self.model)
# get predictions
if U.is_huggingface(model=self.model):
tseq = self.preproc.preprocess_test(texts, verbose=0)
tseq.batch_size = self.batch_size
tfd = tseq.to_tfdataset(train=False)
preds = self.model.predict(tfd)
if type(preds).__name__ == 'TFSequenceClassifierOutput': # dep_fix: undocumented breaking change in transformers==4.0.0
preds = preds.logits
# dep_fix: transformers in TF 2.2.0 returns a tuple insead of NumPy array for some reason
if isinstance(preds, tuple) and len(preds) == 1: preds = preds[0]
else:
texts = self.preproc.preprocess(texts)
preds = self.model.predict(texts, batch_size=self.batch_size)
# process predictions
if U.is_huggingface(model=self.model):
# convert logits to probabilities for Hugging Face models
if multilabel and self.c:
preds = activations.sigmoid(tf.convert_to_tensor(preds)).numpy()
elif self.c:
preds = activations.softmax(tf.convert_to_tensor(preds)).numpy()
else:
preds = np.squeeze(preds)
if len(preds.shape) == 0: preds = np.expand_dims(preds, -1)
result = preds if return_proba or multilabel or not self.c else [self.c[np.argmax(pred)] for pred in preds]
if multilabel and not return_proba:
result = [list(zip(self.c, r)) for r in result]
if not is_array: return result[0]
else: return result
def predict_proba(self, texts):
"""
```
Makes predictions for a list of strings where each string is a document
or text snippet.
Returns probabilities of each class.
```
"""
return self.predict(texts, return_proba=True)
def explain(self, doc, truncate_len=512, all_targets=False, n_samples=2500):
"""
Highlights text to explain prediction
Args:
doc (str): text of documnet
truncate_len(int): truncate document to this many words
all_targets(bool): If True, show visualization for
each target.
n_samples(int): number of samples to generate and train on.
Larger values give better results, but will take more time.
Lower this value if explain is taking too long.
"""
is_array, is_pair = detect_text_format(doc)
if is_pair:
warnings.warn('currently_unsupported: explain does not currently support sentence pair classification')
return
if not self.c:
warnings.warn('currently_unsupported: explain does not support text regression')
return
try:
import eli5
from eli5.lime import TextExplainer
except:
msg = 'ktrain requires a forked version of eli5 to support tf.keras. '+\
'Install with: pip install https://github.com/amaiya/eli5/archive/refs/heads/tfkeras_0_10_1.zip'
warnings.warn(msg)
return
if not hasattr(eli5, 'KTRAIN_ELI5_TAG') or eli5.KTRAIN_ELI5_TAG != KTRAIN_ELI5_TAG:
msg = 'ktrain requires a forked version of eli5 to support tf.keras. It is either missing or not up-to-date. '+\
'Uninstall the current version and install/re-install the fork with: pip install https://github.com/amaiya/eli5/archive/refs/heads/tfkeras_0_10_1.zip'
warnings.warn(msg)
return
if not isinstance(doc, str): raise TypeError('text must of type str')
prediction = [self.predict(doc)] if not all_targets else None
if self.preproc.is_nospace_lang():
doc = self.preproc.process_chinese([doc])
doc = doc[0]
doc = ' '.join(doc.split()[:truncate_len])
te = TextExplainer(random_state=42, n_samples=n_samples)
_ = te.fit(doc, self.predict_proba)
return te.show_prediction(target_names=self.preproc.get_classes(), targets=prediction)
def _save_model(self, fpath):
if isinstance(self.preproc, TransformersPreprocessor):
self.model.save_pretrained(fpath)
# As of 0.26.3, make sure we save tokenizer in predictor folder
tok = self.preproc.get_tokenizer()
tok.save_pretrained(fpath)
else:
super()._save_model(fpath)
return
def export_model_to_onnx(self, fpath, quantize=False, target_opset=None, verbose=1):
"""
```
Export model to onnx
Args:
fpath(str): String representing full path to model file where ONNX model will be saved.
Example: '/tmp/my_model.onnx'
quantize(str): If True, will create a total of three model files will be created using transformers.convert_graph_to_onnx:
1) ONNX model (created directly using keras2onnx
2) an optimized ONNX model (created by transformers library)
3) a quantized version of optimized ONNX model (created by transformers library)
All files will be created in the parent folder of fpath:
Example:
If fpath='/tmp/model.onnx', then both /tmp/model-optimized.onnx and
/tmp/model-optimized-quantized.onnx will also be created.
verbose(bool): verbosity
Returns:
str: string representing fpath. If quantize=True, returned fpath will be different than supplied fpath
```
"""
try:
import onnxruntime, onnxruntime_tools, onnx, keras2onnx
except ImportError:
raise Exception('This method requires ONNX libraries to be installed: '+\
'pip install -q --upgrade onnxruntime==1.5.1 onnxruntime-tools onnx keras2onnx')
from pathlib import Path
if type(self.preproc).__name__ == 'BERTPreprocessor':
raise Exception('currently_unsupported: BERT models created with text_classifier("bert",...) are not supported (i.e., keras_bert models). ' +\
'Only BERT models created with Transformer(...) are supported.')
if verbose: print('converting to ONNX format ... this may take a few moments...')
if U.is_huggingface(model=self.model):
tokenizer = self.preproc.get_tokenizer()
maxlen = self.preproc.maxlen
input_dict = tokenizer('Name', return_tensors='tf',
padding='max_length', max_length=maxlen)
if version.parse(tf.__version__) < version.parse('2.2'):
raise Exception('export_model_to_tflite requires tensorflow>=2.2')
#self.model._set_inputs(input_spec, training=False) # for tf < 2.2
self.model._saved_model_inputs_spec = None # for tf > 2.2
self.model._set_save_spec(input_dict) # for tf > 2.2
self.model._get_save_spec()
onnx_model = keras2onnx.convert_keras(self.model, self.model.name, target_opset=target_opset)
keras2onnx.save_model(onnx_model, fpath)
return_fpath = fpath
if quantize:
from transformers.convert_graph_to_onnx import optimize, quantize
#opt_path = optimize(Path(fpath))
if U.is_huggingface(model=self.model) and\
type(self.model).__name__ in ['TFDistilBertForSequenceClassification', 'TFBertForSequenceClassification']:
try:
from onnxruntime_tools import optimizer
from onnxruntime_tools.transformers.onnx_model_bert import BertOptimizationOptions
# disable embedding layer norm optimization for better model size reduction
opt_options = BertOptimizationOptions('bert')
opt_options.enable_embed_layer_norm = False
opt_model = optimizer.optimize_model(
fpath,
'bert', # bert_keras causes error with transformers
num_heads=12,
hidden_size=768,
optimization_options=opt_options)
opt_model.save_model_to_file(fpath)
except:
warnings.warn('Could not run BERT-specific optimizations')
pass
quantize_path = quantize(Path(fpath))
return_fpath = quantize_path.as_posix()
if verbose: print('done.')
return return_fpath</code></pre>
</details>
</section>
<section>
</section>
<section>
</section>
<section>
</section>
<section>
<h2 class="section-title" id="header-classes">Classes</h2>
<dl>
<dt id="ktrain.text.predictor.TextPredictor"><code class="flex name class">
<span>class <span class="ident">TextPredictor</span></span>
<span>(</span><span>model, preproc, batch_size=32)</span>
</code></dt>
<dd>
<div class="desc"><pre><code>predicts text classes
</code></pre></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">class TextPredictor(Predictor):
"""
```
predicts text classes
```
"""
def __init__(self, model, preproc, batch_size=U.DEFAULT_BS):
if not isinstance(model, Model):
raise ValueError('model must be of instance Model')
if not isinstance(preproc, TextPreprocessor):
#if type(preproc).__name__ != 'TextPreprocessor':
raise ValueError('preproc must be a TextPreprocessor object')
self.model = model
self.preproc = preproc
self.c = self.preproc.get_classes()
self.batch_size = batch_size
def get_classes(self):
return self.c
def predict(self, texts, return_proba=False):
"""
```
Makes predictions for a list of strings where each string is a document
or text snippet.
If return_proba is True, returns probabilities of each class.
Args:
texts(str|list): For text classification, texts should be either a str or
a list of str.
For sentence pair classification, texts should be either
a tuple of form (str, str) or list of tuples.
A single tuple of the form (str, str) is automatically treated as sentence pair classification, so
please refrain from using tuples for text classification tasks.
return_proba(bool): If True, return probabilities instead of predicted class labels
```
"""
is_array, is_pair = detect_text_format(texts)
if not is_array: texts = [texts]
classification, multilabel = U.is_classifier(self.model)
# get predictions
if U.is_huggingface(model=self.model):
tseq = self.preproc.preprocess_test(texts, verbose=0)
tseq.batch_size = self.batch_size
tfd = tseq.to_tfdataset(train=False)
preds = self.model.predict(tfd)
if type(preds).__name__ == 'TFSequenceClassifierOutput': # dep_fix: undocumented breaking change in transformers==4.0.0
preds = preds.logits
# dep_fix: transformers in TF 2.2.0 returns a tuple insead of NumPy array for some reason
if isinstance(preds, tuple) and len(preds) == 1: preds = preds[0]
else:
texts = self.preproc.preprocess(texts)
preds = self.model.predict(texts, batch_size=self.batch_size)
# process predictions
if U.is_huggingface(model=self.model):
# convert logits to probabilities for Hugging Face models
if multilabel and self.c:
preds = activations.sigmoid(tf.convert_to_tensor(preds)).numpy()
elif self.c:
preds = activations.softmax(tf.convert_to_tensor(preds)).numpy()
else:
preds = np.squeeze(preds)
if len(preds.shape) == 0: preds = np.expand_dims(preds, -1)
result = preds if return_proba or multilabel or not self.c else [self.c[np.argmax(pred)] for pred in preds]
if multilabel and not return_proba:
result = [list(zip(self.c, r)) for r in result]
if not is_array: return result[0]
else: return result
def predict_proba(self, texts):
"""
```
Makes predictions for a list of strings where each string is a document
or text snippet.
Returns probabilities of each class.
```
"""
return self.predict(texts, return_proba=True)
def explain(self, doc, truncate_len=512, all_targets=False, n_samples=2500):
"""
Highlights text to explain prediction
Args:
doc (str): text of documnet
truncate_len(int): truncate document to this many words
all_targets(bool): If True, show visualization for
each target.
n_samples(int): number of samples to generate and train on.
Larger values give better results, but will take more time.
Lower this value if explain is taking too long.
"""
is_array, is_pair = detect_text_format(doc)
if is_pair:
warnings.warn('currently_unsupported: explain does not currently support sentence pair classification')
return
if not self.c:
warnings.warn('currently_unsupported: explain does not support text regression')
return
try:
import eli5
from eli5.lime import TextExplainer
except:
msg = 'ktrain requires a forked version of eli5 to support tf.keras. '+\
'Install with: pip install https://github.com/amaiya/eli5/archive/refs/heads/tfkeras_0_10_1.zip'
warnings.warn(msg)
return
if not hasattr(eli5, 'KTRAIN_ELI5_TAG') or eli5.KTRAIN_ELI5_TAG != KTRAIN_ELI5_TAG:
msg = 'ktrain requires a forked version of eli5 to support tf.keras. It is either missing or not up-to-date. '+\
'Uninstall the current version and install/re-install the fork with: pip install https://github.com/amaiya/eli5/archive/refs/heads/tfkeras_0_10_1.zip'
warnings.warn(msg)
return
if not isinstance(doc, str): raise TypeError('text must of type str')
prediction = [self.predict(doc)] if not all_targets else None
if self.preproc.is_nospace_lang():
doc = self.preproc.process_chinese([doc])
doc = doc[0]
doc = ' '.join(doc.split()[:truncate_len])
te = TextExplainer(random_state=42, n_samples=n_samples)
_ = te.fit(doc, self.predict_proba)
return te.show_prediction(target_names=self.preproc.get_classes(), targets=prediction)
def _save_model(self, fpath):
if isinstance(self.preproc, TransformersPreprocessor):
self.model.save_pretrained(fpath)
# As of 0.26.3, make sure we save tokenizer in predictor folder
tok = self.preproc.get_tokenizer()
tok.save_pretrained(fpath)
else:
super()._save_model(fpath)
return
def export_model_to_onnx(self, fpath, quantize=False, target_opset=None, verbose=1):
"""
```
Export model to onnx
Args:
fpath(str): String representing full path to model file where ONNX model will be saved.
Example: '/tmp/my_model.onnx'
quantize(str): If True, will create a total of three model files will be created using transformers.convert_graph_to_onnx:
1) ONNX model (created directly using keras2onnx
2) an optimized ONNX model (created by transformers library)
3) a quantized version of optimized ONNX model (created by transformers library)
All files will be created in the parent folder of fpath:
Example:
If fpath='/tmp/model.onnx', then both /tmp/model-optimized.onnx and
/tmp/model-optimized-quantized.onnx will also be created.
verbose(bool): verbosity
Returns:
str: string representing fpath. If quantize=True, returned fpath will be different than supplied fpath
```
"""
try:
import onnxruntime, onnxruntime_tools, onnx, keras2onnx
except ImportError:
raise Exception('This method requires ONNX libraries to be installed: '+\
'pip install -q --upgrade onnxruntime==1.5.1 onnxruntime-tools onnx keras2onnx')
from pathlib import Path
if type(self.preproc).__name__ == 'BERTPreprocessor':
raise Exception('currently_unsupported: BERT models created with text_classifier("bert",...) are not supported (i.e., keras_bert models). ' +\
'Only BERT models created with Transformer(...) are supported.')
if verbose: print('converting to ONNX format ... this may take a few moments...')
if U.is_huggingface(model=self.model):
tokenizer = self.preproc.get_tokenizer()
maxlen = self.preproc.maxlen
input_dict = tokenizer('Name', return_tensors='tf',
padding='max_length', max_length=maxlen)
if version.parse(tf.__version__) < version.parse('2.2'):
raise Exception('export_model_to_tflite requires tensorflow>=2.2')
#self.model._set_inputs(input_spec, training=False) # for tf < 2.2
self.model._saved_model_inputs_spec = None # for tf > 2.2
self.model._set_save_spec(input_dict) # for tf > 2.2
self.model._get_save_spec()
onnx_model = keras2onnx.convert_keras(self.model, self.model.name, target_opset=target_opset)
keras2onnx.save_model(onnx_model, fpath)
return_fpath = fpath
if quantize:
from transformers.convert_graph_to_onnx import optimize, quantize
#opt_path = optimize(Path(fpath))
if U.is_huggingface(model=self.model) and\
type(self.model).__name__ in ['TFDistilBertForSequenceClassification', 'TFBertForSequenceClassification']:
try:
from onnxruntime_tools import optimizer
from onnxruntime_tools.transformers.onnx_model_bert import BertOptimizationOptions
# disable embedding layer norm optimization for better model size reduction
opt_options = BertOptimizationOptions('bert')
opt_options.enable_embed_layer_norm = False
opt_model = optimizer.optimize_model(
fpath,
'bert', # bert_keras causes error with transformers
num_heads=12,
hidden_size=768,
optimization_options=opt_options)
opt_model.save_model_to_file(fpath)
except:
warnings.warn('Could not run BERT-specific optimizations')
pass
quantize_path = quantize(Path(fpath))
return_fpath = quantize_path.as_posix()
if verbose: print('done.')
return return_fpath</code></pre>
</details>
<h3>Ancestors</h3>
<ul class="hlist">
<li><a title="ktrain.predictor.Predictor" href="../predictor.html#ktrain.predictor.Predictor">Predictor</a></li>
<li>abc.ABC</li>
</ul>
<h3>Methods</h3>
<dl>
<dt id="ktrain.text.predictor.TextPredictor.explain"><code class="name flex">
<span>def <span class="ident">explain</span></span>(<span>self, doc, truncate_len=512, all_targets=False, n_samples=2500)</span>
</code></dt>
<dd>
<div class="desc"><p>Highlights text to explain prediction</p>
<h2 id="args">Args</h2>
<dl>
<dt><strong><code>doc</code></strong> : <code>str</code></dt>
<dd>text of documnet</dd>
</dl>
<p>truncate_len(int): truncate document to this many words
all_targets(bool):
If True, show visualization for
each target.
n_samples(int): number of samples to generate and train on.
Larger values give better results, but will take more time.
Lower this value if explain is taking too long.</p></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">def explain(self, doc, truncate_len=512, all_targets=False, n_samples=2500):
"""
Highlights text to explain prediction
Args:
doc (str): text of documnet
truncate_len(int): truncate document to this many words
all_targets(bool): If True, show visualization for
each target.
n_samples(int): number of samples to generate and train on.
Larger values give better results, but will take more time.
Lower this value if explain is taking too long.
"""
is_array, is_pair = detect_text_format(doc)
if is_pair:
warnings.warn('currently_unsupported: explain does not currently support sentence pair classification')
return
if not self.c:
warnings.warn('currently_unsupported: explain does not support text regression')
return
try:
import eli5
from eli5.lime import TextExplainer
except:
msg = 'ktrain requires a forked version of eli5 to support tf.keras. '+\
'Install with: pip install https://github.com/amaiya/eli5/archive/refs/heads/tfkeras_0_10_1.zip'
warnings.warn(msg)
return
if not hasattr(eli5, 'KTRAIN_ELI5_TAG') or eli5.KTRAIN_ELI5_TAG != KTRAIN_ELI5_TAG:
msg = 'ktrain requires a forked version of eli5 to support tf.keras. It is either missing or not up-to-date. '+\
'Uninstall the current version and install/re-install the fork with: pip install https://github.com/amaiya/eli5/archive/refs/heads/tfkeras_0_10_1.zip'
warnings.warn(msg)
return
if not isinstance(doc, str): raise TypeError('text must of type str')
prediction = [self.predict(doc)] if not all_targets else None
if self.preproc.is_nospace_lang():
doc = self.preproc.process_chinese([doc])
doc = doc[0]
doc = ' '.join(doc.split()[:truncate_len])
te = TextExplainer(random_state=42, n_samples=n_samples)
_ = te.fit(doc, self.predict_proba)
return te.show_prediction(target_names=self.preproc.get_classes(), targets=prediction)</code></pre>
</details>
</dd>
<dt id="ktrain.text.predictor.TextPredictor.get_classes"><code class="name flex">
<span>def <span class="ident">get_classes</span></span>(<span>self)</span>
</code></dt>
<dd>
<div class="desc"></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">def get_classes(self):
return self.c</code></pre>
</details>
</dd>
<dt id="ktrain.text.predictor.TextPredictor.predict"><code class="name flex">
<span>def <span class="ident">predict</span></span>(<span>self, texts, return_proba=False)</span>
</code></dt>
<dd>
<div class="desc"><pre><code>
Makes predictions for a list of strings where each string is a document
or text snippet.
If return_proba is True, returns probabilities of each class.
Args:
texts(str|list): For text classification, texts should be either a str or
a list of str.
For sentence pair classification, texts should be either
a tuple of form (str, str) or list of tuples.
A single tuple of the form (str, str) is automatically treated as sentence pair classification, so
please refrain from using tuples for text classification tasks.
return_proba(bool): If True, return probabilities instead of predicted class labels
</code></pre></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">def predict(self, texts, return_proba=False):
"""
```
Makes predictions for a list of strings where each string is a document
or text snippet.
If return_proba is True, returns probabilities of each class.
Args:
texts(str|list): For text classification, texts should be either a str or
a list of str.
For sentence pair classification, texts should be either
a tuple of form (str, str) or list of tuples.
A single tuple of the form (str, str) is automatically treated as sentence pair classification, so
please refrain from using tuples for text classification tasks.
return_proba(bool): If True, return probabilities instead of predicted class labels
```
"""
is_array, is_pair = detect_text_format(texts)
if not is_array: texts = [texts]
classification, multilabel = U.is_classifier(self.model)
# get predictions
if U.is_huggingface(model=self.model):
tseq = self.preproc.preprocess_test(texts, verbose=0)
tseq.batch_size = self.batch_size
tfd = tseq.to_tfdataset(train=False)
preds = self.model.predict(tfd)
if type(preds).__name__ == 'TFSequenceClassifierOutput': # dep_fix: undocumented breaking change in transformers==4.0.0
preds = preds.logits
# dep_fix: transformers in TF 2.2.0 returns a tuple insead of NumPy array for some reason
if isinstance(preds, tuple) and len(preds) == 1: preds = preds[0]
else:
texts = self.preproc.preprocess(texts)
preds = self.model.predict(texts, batch_size=self.batch_size)
# process predictions
if U.is_huggingface(model=self.model):
# convert logits to probabilities for Hugging Face models
if multilabel and self.c:
preds = activations.sigmoid(tf.convert_to_tensor(preds)).numpy()
elif self.c:
preds = activations.softmax(tf.convert_to_tensor(preds)).numpy()
else:
preds = np.squeeze(preds)
if len(preds.shape) == 0: preds = np.expand_dims(preds, -1)
result = preds if return_proba or multilabel or not self.c else [self.c[np.argmax(pred)] for pred in preds]
if multilabel and not return_proba:
result = [list(zip(self.c, r)) for r in result]
if not is_array: return result[0]
else: return result</code></pre>
</details>
</dd>
<dt id="ktrain.text.predictor.TextPredictor.predict_proba"><code class="name flex">
<span>def <span class="ident">predict_proba</span></span>(<span>self, texts)</span>
</code></dt>
<dd>
<div class="desc"><pre><code>Makes predictions for a list of strings where each string is a document
or text snippet.
Returns probabilities of each class.
</code></pre></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">def predict_proba(self, texts):
"""
```
Makes predictions for a list of strings where each string is a document
or text snippet.
Returns probabilities of each class.
```
"""
return self.predict(texts, return_proba=True)</code></pre>
</details>
</dd>
</dl>
<h3>Inherited members</h3>
<ul class="hlist">
<li><code><b><a title="ktrain.predictor.Predictor" href="../predictor.html#ktrain.predictor.Predictor">Predictor</a></b></code>:
<ul class="hlist">
<li><code><a title="ktrain.predictor.Predictor.create_onnx_session" href="../predictor.html#ktrain.predictor.Predictor.create_onnx_session">create_onnx_session</a></code></li>
<li><code><a title="ktrain.predictor.Predictor.export_model_to_onnx" href="../predictor.html#ktrain.predictor.Predictor.export_model_to_onnx">export_model_to_onnx</a></code></li>
<li><code><a title="ktrain.predictor.Predictor.export_model_to_tflite" href="../predictor.html#ktrain.predictor.Predictor.export_model_to_tflite">export_model_to_tflite</a></code></li>
<li><code><a title="ktrain.predictor.Predictor.save" href="../predictor.html#ktrain.predictor.Predictor.save">save</a></code></li>
</ul>
</li>
</ul>
</dd>
</dl>
</section>
</article>
<nav id="sidebar">
<h1>Index</h1>
<div class="toc">
<ul></ul>
</div>
<ul id="index">
<li><h3>Super-module</h3>
<ul>
<li><code><a title="ktrain.text" href="index.html">ktrain.text</a></code></li>
</ul>
</li>
<li><h3><a href="#header-classes">Classes</a></h3>
<ul>
<li>
<h4><code><a title="ktrain.text.predictor.TextPredictor" href="#ktrain.text.predictor.TextPredictor">TextPredictor</a></code></h4>
<ul class="">
<li><code><a title="ktrain.text.predictor.TextPredictor.explain" href="#ktrain.text.predictor.TextPredictor.explain">explain</a></code></li>
<li><code><a title="ktrain.text.predictor.TextPredictor.get_classes" href="#ktrain.text.predictor.TextPredictor.get_classes">get_classes</a></code></li>
<li><code><a title="ktrain.text.predictor.TextPredictor.predict" href="#ktrain.text.predictor.TextPredictor.predict">predict</a></code></li>
<li><code><a title="ktrain.text.predictor.TextPredictor.predict_proba" href="#ktrain.text.predictor.TextPredictor.predict_proba">predict_proba</a></code></li>
</ul>
</li>
</ul>
</li>
</ul>
</nav>
</main>
<footer id="footer">
<p>Generated by <a href="https://pdoc3.github.io/pdoc"><cite>pdoc</cite> 0.9.2</a>.</p>
</footer>
</body>
</html>